• Journal of Korea Multimedia Society
• /
• v.24 no.4
• /
• pp.550-557
• /
• 2021

Image steganalysis is a technique for detecting images with steganographic algorithms applied, called stego images. With state-of-the-art CNN-based steganalysis methods, we can detect stego images with high accuracy, but it is not possible to know which steganographic algorithm is used. Identifying stego images is essential for extracting embedded data. In this paper, as the first step for extracting data from stego images, we propose a hierarchical CNN structure for senary classification of steganographic algorithms. The hierarchical CNN structure consists of multiple CNN networks which are trained to classify each steganographic algorithm and performs binary or ternary classification. Thus, it classifies multiple steganogrphic algorithms hierarchically and stepwise, rather than classifying them at the same time. In experiments of comparing with several conventional methods, including those of classifying multiple steganographic algorithms at the same time, it is verified that using the hierarchical CNN structure can greatly improve the classification accuracy.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.540-543
• /
• 2020

기존의 초해상도 딥러닝 기법은 모델의 깊이가 깊어지면서, 좋은 성능을 내지만 점점 더 복잡해지고 있고, 실제로 사용하는데 있어 많은 시간을 요구한다. 이를 해결하기 위해, 우리는 딥러닝 모델의 가중치를 양자화 하여 추론시간을 줄이고자 한다. 초해상도 모델은 feature extraction, non-linear mapping, reconstruction 세 부분으로 나누어져 있으며, 레이어 사이에 많은 skip-connection 이 존재하는 특징이 있다. 따라서 양자화 시 최종 성능 하락에 미치는 영향력이 레이어 별로 다르며, 이를 감안하여 강화학습으로 레이어 별 최적 bit 를 찾아 성능 하락을 최소화한다. 본 논문에서는 Skip-connection 이 많이 존재하는 MSRN 을 사용하였으며, 결과에서 feature extraction, reconstruction 부분과 블록 내 특정 위치의 레이어가 항상 높은 bit 를 가짐을 알 수 있다. 기존에 영상 분류에 한정되어 사용되었던 혼합 bit 양자화를 사용하여 초해상도 딥러닝 기법의 모델 사이즈를 줄인 최초의 논문이며, 제안 방법은 모바일 등 제한된 환경에 적용 가능할 것으로 생각된다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1311-1314
• /
• 2022

DNN을 사용하여 객체 인식 과정에서 객체를 잘 분류하기 위해서는 시각적 설명성이 요구된다. 시각적 설명성은 object class에 대한 예측을 pixel-wise attribution으로 표현해 예측 근거를 해석하기 위해 제안되었다, Scale-invariant한 특징을 제공하도록 설계된 pyramidal features 기반 backbone 구조는 object detection 및 classification 등에서 널리 쓰이고 있으며, 이러한 특징을 갖는 feature pyramid를 trainable attention mechanism에 적용하고자 할 때 계산량 및 메모리의 복잡도가 증가하는 문제가 있다. 본 논문에서는 일반적인 FPN에서 객체 인식 성능과 설명성을 높이기 위한 피라미드-주의집중 계층네트워크 (FPN-Attention Layered Network) 방식을 제안하고, 실험적으로 그 특성을 평가하고자 한다. 기존의 FPN만을 사용하였을 때 객체 인식 과정에서 설명성을 향상시키는 방식이 객체 인식에 미치는 정도를 정량적으로 평가하였다. 제안된 모델의 적용을 통해 낮은 computing 오버헤드 수준에서 multi-level feature를 고려한 시각적 설명성을 개선시켜, 결괴적으로 객체 인식 성능을 향상 시킬 수 있음을 실험적으로 확인할 수 있었다.

• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2022.10a
• /
• pp.150-151
• /
• 2022

본 논문에서는 대규모 평시 또는 감염병 발생이전 지자체별 취약성 평가를 통해 감염병 발생 및 확산에 영향을 미치는 취약지표를 평가해보는 것이다. 이를 위해 국내 코로나19 대처사항(각종 회의자료 및 보고자료 등), 선행연구 등을 기반으로 확산취약시설, 감염취약계층, 사회적 취약여건, 대응능력(보건·의료/재정적, 사회적 대응역량) 총 4개의 증감유형으로 분류한 후 세부 평가지표로 약 24개로 세부지표를 선정하였다. 이러한 취약성 평가 결과는 지자체별 특성에 맞는 감염병 대응정책 수립에 필요한 근거자료를 마련하는 기초자료로 활용 가능할 것이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.828-831
• /
• 2008

플래시 메모리(Flash Memory) 기술이 빠르게 발전하면서, 플래시 메모리 기반의 저장 장치가 개인용 컴퓨터나 엔터프라이즈 서버 시스템과 같은 시스템에 2차적인 저장 장치로써 사용가능해지고 있다. FTL(Flash Translation Layer)의 기본적인 기능은 플래시 메모리의 논리 주소를 물리 주소로 바꾸는 것임에도 불구하고, FTL의 효율적인 알고리즘은 성능과 수명에 상당한 효과를 가지고 있다. 이 논문에서는 MP3 플레이어와 디지털 카메라, SSDs(Solid-State Disk)와 같은 낸드 플래시 메모리(NAND Flash Memory) 기반의 어플리케이션을 위한 N : N+K 매핑을 사용하는 새로운 FTL 설계를 제안한다. 성능에 영향을 미치는 매개변수들을 분류하여, 다양한 워크로드 분석을 기반으로 FTL을 조사했다. 우리가 제안하는 FTL을 가지고, 낸드 플래시 어플리케이션 가동에 따라 어떤 매개변수가 최대 성능을 낼 수 있는지 알아낼 수 있고, 그 변수들을 유연하게 조정하여 성능을 더 향상시킬 수 있다.

• Korean Journal of Microbiology
• /
• v.49 no.2
• /
• pp.150-155
• /
• 2013

We have conducted pyrosequencing for freshwater microbial community analyses. Fourteen sites along the Yeongsan river were selected for this study, and samples were collected monthly from May to July, 2012. Total 987,380 reads were obtained from 42 samples and used for taxonomic classification and OTU distribution analysis. Our results showed that high geographical and temporal variation in the phylum level bacterial composition, suggesting that microbial community is a very sensitive parameter affected by the surrounding environments including tributaries and land use nearby. In addition, we conducted an OTU-based Microbial Source Tracking to identify sources of fecal pollution in the same region. From this study Firmicutes was found to be the most influential taxa in this region. Here, we report that the use of pyrosequencing based microbial community analysis may give an additional information on freshwater quality monitoring, in addition to the currently used water quality parameters, such as BOD and pH.

• The Journal of the Korea Contents Association
• /
• v.9 no.12
• /
• pp.814-825
• /
• 2009

This paper is to analyze the relative weight and the priority order of the selection factors for IPTV. To obtain the goal this paper, firstly, the selecting factors for IPTV were identified and conceptualized on reviewing previous literatures. And this paper proposed the hierarchy model using the factor analysis on IPTV users. and utilized AHP in analysis method. Secondly, AHP model constructed 4 higher factors, and 15 lower factors. The higher factors were 'Quality Acceptance', 'Social Effect', 'Using Motive' and 'Individual Psychology'. Results of this study show that relative weights among factors IPTV selection were founded as 'Quality Acceptance(33.0%)', 'Utilization Motive(31.8%)', 'Individual Psychology(21.3%)' and 'Social Influence(13.9%)'. Economic efficiency takes the top rank in sub-criteria included 15 factors and than convenience, amount of information, discrimination, self-efficacy, innovation, relative advantage in order listed. For the generalization of the research results, further researchs are needed to investigate intended use of IPTV on other method of analysis.

• Journal of the Korean Society of Groundwater Environment
• /
• v.4 no.3
• /
• pp.122-129
• /
• 1997

Spills or leaks of hazardous organic. compounds from underground storage tanks (USTs) are common contaminant sources of soil and groundwater. It would aid in managing USTs to assess and rank the potential environmental hazard posed by the USTs. Therefore, a preliminary hazard ranking system of USTs is developed in this study. The system is combined with GIS and consists of five steps: 1) selection of significant factors, 2) determination of the hierarchy of the factors. 3) determination of the weights, 4) calculation of the potential hazard, and 5) hazard assessment. The system is applied to the gas stations in Kwanak-gu, Seoul. The results indicate that the gas stations can be categorized in three groups as highly hazardous, less highly hazardous, and weakly hazardous. Seven gas stations belong to the highly hazardous group. Through the sensitivity analysis, four stations appear to possess high hazard potentials regardless of weights assigned to the factors. It appears that a user can make flexible application of the hazard ranking system with the user's experience and particular purposes. However, the system still needs validations against field survey data.

• The Journal of Engineering Geology
• /
• v.27 no.1
• /
• pp.1-7
• /
• 2017

Seawater intrusion was evaluated using electrical conductivity, chemical proxies, and oxygen isotopic compositions in coastal area. It seems that groundwater in the area where the electrical conductivity is over $3,000{\mu}S/cm$ is influenced by seawater. However, it is very difficult that seawater intrusion is distinguished from other contaminants using the electrical conductivity. The chemical proxies and oxygen and hydrogen isotopic compositions can quantitatively estimate seawater intrusion. However, these method is a costly disadvantage. Therefore, firstly, groundwater contamination by seawater intrusion was evaluated using electrical conductivity and then the additional chemical and isotopic analysis were conducted in areas where possibility of contamination by seawater intrusion is high. These systematic analytical method can reduce analytical cost to quantitatively evaluate influence of seawater intrusion on coastal groundwater and may improve efficiency of analytical method for seawater intrusion.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.23 no.1
• /
• pp.57-62
• /
• 2011

At this point that port industry brings a new change as the concept of 'Green Logistics' appears, studies on an economic analysis of existing port Re-modeling have normally been conducted focused on market value factors such as Benefit-Cost analysis, On the other hand, there have been few studies on the analysis of Non-market value factors such as environmental value. Therefore, the purpose of this study is aimed to establish a Feasibility analysis model of Incheon inner port Re-modeling in the aspect of Non-market Benefit Factors classified by related factors such as, marine environment, waterfront facility and welfare of users by using AHP methodology based on the extracted result through previous studies.